The majority of the genes that our group has explored, like GRM3, DISC1, NRG1, ERBB4 and KCHN2, show alternative splicing in postmortem human brain and the mechanism of genetic association in many cases appears to relate specifically to regulation of expression of novel splice forms. Many of the schizophrenia associated splice variants appear to be abundantly expressed in fetal brain, implicating early brain development as a critical time for the molecular effects of risk associated genetic variation. Our group continues to develop expression profile datasets using commercially available platforms and microarray chip SNP genotyping for the identification of common variations associated with gene expression. This has proven to be an invaluable and unique resource for not only validation of genes, showing association with schizophrenia and related intermediate phenotypes, but also for identifying potentially critical gene modifiers of various central nervous system disorders, including neurodegeneration and ageing. An example of this validation process is our study of metabolic glutamate receptor 2 and 3 gene expression. Previous research data have shown that in both rodent and human, Group II metabotropic glutamate receptors (mGluR2 and mGluR3) are associated in the pathophysiology of schizophrenia. Rodents given PCP, model schizophrenia ,with increased glutamate in prefrontal cortex (PFC), increased motor activity and movement, and impaired performance on working memory tasks can be prevented with pretreatment of an mGluR binding drug (LY 404039). Likewise, in humans, ketamine has been shown to induce cognitive deficits similar to those found in schizophrenia. Again pretreatment with the mGluR agonist LY 404039 reduced the cognitive deficits caused by the ketamine. Increased levels of mGluR2 and -3 have been reported in the PFC in schizophrenia and LY 404039 has been reported to have antipsychotic activity in patients with schizophrenia. The expression of these receptors had not been characterized in the human brain. We sought to characterize the normal distribution of mGluR2 and -3 in human PFC and mesencephalon and compare schizophrenia with matched controls. We found that mGluR2 and -3 mRNA is expressed in human PFC in dopamine cells. The expression of these receptors in dopamine cells provides a mechanism for glutamate to modulate dopamine release in the human brain. This is a species-specific difference which may be critical in understanding the rodent models in schizophrenia. More importantly, this species-specific distribution patter of mGluR2 and -3 in dopamine neurons suggests the presence of these receptors in projections form the PFC and mesencephalon. The circuit involving the mesencephalon, striatum, and PFC has been implicated as the psychosis circuit. We also found significantly higher levels of mGluR2 mRNA in the PFC white matter in schizophrenia versus matched controls, which is consistent with the evidence of white matter abnormalities we find in schizophrenia. To date, there are no reports of normal distribution of mGluR2 mRNA in human PFC.